A multi-scale intelligent simulation method for geothermal development scenarios based on hierarchical coupling

Abstract

The present invention provides a multi-scale intelligent simulation method for geothermal development scenarios based on hierarchical coupling, belonging to the field of intelligent simulation. The method includes: constructing multiple-scale models corresponding to geothermal resource data according to the geothermal development scenario; constructing a multi-scale simulation model, and using a deep neural network to perform deep feature learning on the multi-scale simulation model to obtain a deep feature representation; according to the deep feature representation of the multi-scale simulation model, obtaining a common subspace through feature fusion, and performing feature balancing and feature alignment to obtain a hierarchical coupling multi-scale simulation model; conducting simulation experiment verification on the hierarchical coupling multi-scale simulation model, and building a simulation platform for the geothermal development scenario. The present invention solves the problem that the existing simulation methods cannot fully analyze and utilize the deep features of different-scale simulation models themselves, especially the dimensional differences in the feature space may cause a heterogeneous gap, which in turn leads to difficulty in achieving cross-scale feature fusion in the multi-scale framework.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent simulation, and particularly relates to a multi-scale intelligent simulation method for geothermal development scenarios based on hierarchical coupling. Background Art

[0002] Geothermal resources widely exist in the earth's crust and are a clean, stable and renewable energy source. Geothermal development mainly includes exploration, drilling, resource extraction, utilization and management of geothermal resources. Geothermal modeling refers to the mathematical simulation of the formation, distribution, flow, extraction and utilization processes of geothermal resources to optimize development strategies and predict development effects. Traditional modeling methods use numerical methods such as the finite element method and finite difference to solve key numerical values such as temperature and pressure distribution in the geothermal field. Deep learning driven by physical knowledge is an effective tool for computer simulation, but most existing methods are limited to single-scale scenarios described by simple models. During the process of deep geothermal exploitation, high-pressure fluids are injected underground through deep boreholes, forcing micro-nano pores in the target rock formation to expand to several millimeters. If the final formed giant fracture network is considered, the research scope will expand to several kilometers. However, observational and experimental results show that the influence of microdamage accumulation in rocks on fracture behavior often cannot be accurately characterized by macroscopic continuous models. Therefore, how to develop a general cross-scale feature fusion method for efficient communication of multi-scale information is the key to realizing multi-scale simulation by deep learning.

[0003] With the rapid development of artificial intelligence technology, the use of deep learning methods to achieve intelligent simulation of physical problems has received extensive attention from the academic and industrial communities. Currently, among many attempts to apply deep learning methods to multi-scale simulation, it mainly includes research on combining neural networks with traditional numerical methods, multi-scale model coupling methods based on data-driven or dynamic importance sampling, and work on applying transfer learning to simulation.

[0004] However, the above simulation methods simply rely on the knowledge-data comprehensive driving characteristics of deep learning methods, and do not fully analyze and utilize the deep features of different-scale simulation models themselves. Due to the dimensional differences in the feature space, it may cause a heterogeneous gap, which in turn makes it difficult for the multi-scale framework to achieve cross-scale feature fusion; at the same time, there has not yet appeared a method in the field of deep learning simulation that takes into account the deep feature learning of simulation models and effectively completes the hierarchical coupling of multi-scale models. Summary of the Invention

[0005] Aiming at the above deficiencies in the prior art, the present invention provides a multi-scale intelligent simulation method for geothermal development scenarios, which solves the problems that existing simulation methods cannot fully analyze and utilize the deep features of different-scale simulation models themselves, and the dimensional differences in the feature space cause a heterogeneous gap, making it difficult for the multi-scale framework to achieve cross-scale feature fusion.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a multi-scale intelligent simulation method for geothermal development scenarios, comprising the following steps:

[0007] S1. According to the geothermal development scenario, obtain geothermal resource data and construct multiple scale models corresponding to the geothermal resource data;

[0008] S2. Construct a multi-scale simulation model based on multiple scale models, use a deep neural network to perform deep feature learning on the multi-scale simulation model, and obtain a deep feature representation of the multi-scale simulation model;

[0009] S3. Based on the deep feature representation of the multi-scale simulation model, feature fusion is performed on the cross-scale feature space of the multi-scale simulation model to obtain a common subspace, and feature balancing is performed on the common subspace to obtain a hierarchical coupled multi-scale simulation model;

[0010] S4. Based on the physical process, conduct simulation experiments to verify the hierarchical coupled multi-scale simulation model, and use the verified hierarchical coupled multi-scale simulation model to build a simulation platform for geothermal development scenarios to realize multi-scale hierarchical coupled intelligent simulation.

[0011] The beneficial effects of the present invention are as follows: the present invention utilizes physical information neural operators to fully analyze and utilize the deep features of simulation models of different scales themselves, thereby improving the pertinence and feasibility of the simulation of the present invention, and realizing a multi-scale framework that takes into account both deep feature learning of simulation models and cross-scale feature fusion, enriching the existing multi-scale simulation theory; by cross-scale fusion of multi-scale features to obtain a common subspace, the micro-scale model provides missing information to the overall macro-scale model, realizing hierarchical coupling modeling, improving the integrity of feature information, and reducing the dimensionality difference of the feature space.

[0012] Furthermore, the S2 includes the following steps:

[0013] S201, using prior knowledge of multiple scale models of the same physical process and geothermal resource data, pre-training the physical information neural operator to construct an initial multi-scale simulation model;

[0014] S202, performing feature embedding on the initial multi-scale simulation model to obtain a feature-embedded multi-scale simulation model;

[0015] S203. Designing a targeted representation learning model based on the multi-scale simulation model with embedded features, and constructing a deep neural network based on learning and representation of deep features of the multi-scale simulation model based on the targeted representation learning model;

[0016] S204. Use a deep neural network to learn the multi-scale simulation model with feature embeddings, obtain the feature space of the multi-scale simulation model, and obtain the deep feature representation of the multi-scale simulation model related to the geothermal reservoir.

[0017] Furthermore, the expression for feature embedding of the multi-scale simulation model is as follows:

[0018] ;

[0019] Where, represents the predicted value of the function represented by the neural network at the position y , represents the feature embedding of the input function , represents the feature embedding of the input coordinate y , q represents the number of basis functions selected in the target space, represents the neural network 's parameters; where, , , represents q dimensional Euclidean real space.

[0020] The beneficial effect of the above further solution is that the present invention eliminates the semantic gap by designing a targeted representation learning model, constructs a deep neural network to learn and represent the deep features of the pre-trained physics-informed neural operator simulation model, improves the correlation between the deep features and the geothermal reservoir, and lays a solid foundation for the hierarchical coupling of the multi-scale features of the established simulation model feature space.

[0021] Furthermore, the said S3 includes the following steps:

[0022] S301. According to the deep feature representation of the multi-scale simulation model, fuse the cross-scale features of the multi-scale simulation model to obtain a common subspace;

[0023] S302. Use the topological manifold structure of the model data to balance the feature representations under different scale models to obtain a common subspace with balanced features;

[0024] S303. Construct a hierarchical coupled multi-scale simulation model through the transformation matrix between the original high-dimensional features and the low-dimensional common subspace.

[0025] Furthermore, the expression for constructing the hierarchical coupled multi-scale simulation model is as follows:

[0026] ;

[0027] Among them, represents the transformation path, represents the transformation factor, and or 1, represents the transformation matrix from the original high-dimensional feature to the low-dimensional common subspace, represents the transformation matrix from the low-dimensional common subspace to the original high-dimensional feature.

[0028] The beneficial effects of the above further solution are as follows: By using the physical information operator method for developing multi-view feature fusion, the present invention realizes the feature fusion from the feature spaces of different-scale models to the common subspace, introduces the idea of topological manifold, balances the consistency and complementarity between different view representations, improves the robustness of the present invention, and bridges the heterogeneous gap; and through the transformation matrix between the original high-dimensional feature and the low-dimensional common subspace, a hierarchical coupling mechanism of different-scale simulation models is established, realizing the multi-scale simulation of complex physical systems in the form of hierarchical coupling in the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a flowchart of the method of the present invention.

[0030] Figure 2 is an overall structure diagram of the multi-scale intelligent simulation method for geothermal development scenarios based on hierarchical coupling in this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] The following describes the specific embodiments of the present invention to facilitate those skilled in the art to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions made using the concept of the present invention are within the scope of protection.

[0032] Before describing this embodiment, the following terms are first explained:

[0033] PINO: Physical Information Neural Operator;

[0034] K-means algorithm: K-means algorithm.

[0035] Embodiment

[0036] As Figure 1 shown, the present invention provides a multi-scale intelligent simulation method for geothermal development scenarios, and its implementation method is as follows:

[0037] S1. According to the geothermal development scenario, obtain geothermal resource data and construct multiple-scale models corresponding to the geothermal resource data.

[0038] In this embodiment, the deep features of the multi-scale simulation model are related to the long-term evolution of the geothermal reservoir, the cross-scale correlation behavior, and the coupling relationship between different physical fields, etc.

[0039] Such as Figure 2 As shown, obtaining physical models in the form of differential equations at different scales of the geothermal development scenario can be divided into: macroscopic scale models (continuous models based on the conservation laws of quantities such as mass, momentum, and energy), mesoscopic scale models (particle dynamics models simulating the mutual mechanical interactions between rock units), and microscopic scale models (particle models based on first-principles molecular dynamics, quantum mechanics, and Newtonian mechanics, etc.).

[0040] S2. According to multiple scale models, construct a multi-scale simulation model, use a deep neural network to perform deep feature learning on the multi-scale simulation model, and obtain the deep feature representation of the multi-scale simulation model. The specific steps are as follows:

[0041] S201. Use the prior knowledge of the same physical process in multiple scale models and geothermal resource data to pre-train the physics-informed neural operator, and construct an initial multi-scale simulation model;

[0042] S202. Perform feature embedding on the initial multi-scale simulation model to obtain the multi-scale simulation model with feature embedding;

[0043] S203. According to the multi-scale simulation model with feature embedding, design a targeted representation learning model, and based on the targeted representation learning model, construct a deep neural network for learning and representing the deep features of the multi-scale simulation model;

[0044] S204. Use the deep neural network to learn the multi-scale simulation model with feature embedding, learn the feature space of the multi-scale simulation model, and obtain the deep feature representation of the multi-scale simulation model related to the geothermal reservoir.

[0045] In this embodiment, to realize the learning and representation of the deep features of the multi-scale simulation model, use the prior knowledge of the physical models in the form of differential equations of the same physical process at multiple different scales to pre-train the physics-informed neural operator, and establish a simulation model of the corresponding scale, that is, a multi-scale simulation model; according to the process of the physics-informed neural operator model learning the parameterized partial differential equation solver, in the process of constructing the physics-informed neural operator, use the feature embedding based on Fourier features (RFF) to improve the efficiency of the method of the present invention. The calculation formula is as follows:

[0046] ;

[0047] Among them, represents the neural network The function represented at position y has a predicted value of which represents the input function with a feature embedding of which represents the input coordinates y with a feature embedding of q which represents the number of basis functions selected in the target space which represents the neural network parameters; where , , represents q a d-dimensional Euclidean real space; based on automatic differentiation, a differential operator is constructed to minimize the loss function based on partial differential equations , the loss function based on the initial conditions and the loss function based on the boundary conditions to obtain the optimal parameters , completing the optimization of the physics-informed neural operator.

[0048] In this embodiment, the multi-scale simulation model of the present invention is constructed based on the physics-informed neural operator network; a targeted representation learning model for the PINO simulation model is designed, and a deep neural network for learning and representing the deep features of the model is constructed on the basis of eliminating the semantic gap, thereby learning the feature spaces of the PINO simulation model at multiple scales and obtaining the deep feature representation of the multi-scale simulation model related to the geothermal reservoir.

[0049] S3. According to the deep feature representation of the multi-scale simulation model, perform feature fusion on the cross-scale feature space of the multi-scale simulation model to obtain a common subspace, and perform feature balancing on the common subspace to obtain a hierarchical coupled multi-scale simulation model. The specific steps are as follows:

[0050] S301. According to the deep feature representation of the multi-scale simulation model, perform feature fusion on the cross-scale features of the multi-scale simulation model to obtain a common subspace;

[0051] S302. Utilize the topological manifold structure of the model data to balance the feature representations under different scale models to obtain a common subspace with balanced features;

[0052] S303. Construct a hierarchical coupled multi-scale simulation model through the transformation matrix between the original high-dimensional features and the low-dimensional common subspace.

[0053] In this embodiment, based on the analysis of the tensor-based canonical correlation analysis method (TCCA), a fusion from the feature space of the multi-scale model to the same common subspace is studied and constructed, and the preliminary fusion of the features of the multi-scale simulation model is realized; the overall method spans multiple scales of micro, meso, and macro, extracts the key features of each scale, and then integrates the extracted key features through a fusion mechanism to construct a simulation model with global consistency;

[0054] Specifically, as Figure 2 shown, based on the deep feature representation of the multi-scale simulation model, the learned feature representation of the multi-scale simulation model ($ p = 1, 2, …, n; n represents the total number of views, which is the total number of models at different scales in this embodiment), the covariance tensor of all views is calculated to discover the correlation information between all views. By using simplified calculation, it is transformed into an equivalent problem of maximizing the multi-view canonical correlation, that is, finding a set of optimal rank-one tensors ($ k = 1, 2, …, r; r represents the feature dimension after dimensionality reduction, which should be less than the minimum of the feature dimensions of different views), so that the covariance tensor can be expressed as a weighted sum of the rank-one tensors ; the calculation expression is as follows:

[0055] ;

[0056] wherein, represents the tensor product, represents; and the rank-one tensors[[ID=3I]] are integrated into the transformation matrix , and the transformation matrix is applied to the feature representation , and the mapping from the original high-dimensional features to the low-dimensional common subspace is initially realized;

[0057] Furthermore, considering that due to the different physical models used in the simulation models of different scales, there may be large differences in feature representations. Relying solely on maximizing the tensor-based canonical correlation to construct the common subspace cannot completely bridge the heterogeneous gap; after combining the research work on multi-view clustering, the idea of topological manifold is introduced, and the topological manifold structure of the model data is used to balance the consistent structure and complementary information between the feature representations of the simulation models of different scales, thereby improving the robustness of the PINO method for multi-view feature fusion and bridging the heterogeneous gap between the feature spaces of different views.

[0058] In this embodiment, according to the common subspace after balancing the features, as Figure 2As shown, the intelligent simulation model based on the neural operator network at different scales transforms the feature representation through the transformation matrix between the original high-dimensional features and the low-dimensional common subspace, obtaining a hierarchical coupled multi-scale simulation model. The transformation expression is as follows:

[0059] ;

[0060] Among them, represents the transformation path, which is used to define the transformation direction between the original high-dimensional features and the low-dimensional subspace, represents the transformation factor, and or 1, represents the transformation matrix from the original high-dimensional features to the low-dimensional common subspace. When , , the multi-scale simulation model will realize the transformation from the original high-dimensional features to the low-dimensional common subspace, represents the transformation matrix from the low-dimensional common subspace to the original high-dimensional features. When , , the multi-scale simulation model will realize the transformation from the low-dimensional common subspace to the original high-dimensional features.

[0061] S4. Based on the physical process, conduct simulation experiment verification on the hierarchical coupled multi-scale simulation model, and use the verified hierarchical coupled multi-scale simulation model to build a simulation platform for the geothermal development scenario to achieve hierarchical coupled multi-scale intelligent simulation.

[0062] In this embodiment, using the hierarchical coupled multi-scale simulation model, conduct simulation experiment verification based on typical physical processes, take the deep learning of the hierarchical coupled multi-scale simulation model as a multi-scale simulation tool, and build a simulation platform for the geothermal development scenario using the multi-scale simulation tool whose performance has been verified to achieve hierarchical coupled multi-scale simulation.

[0063] In this embodiment, develop the PINO method for hierarchical coupled multi-scale simulation from the deep feature representation of the multi-scale simulation model and the hierarchical coupled multi-scale simulation model.

[0064] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate for implementing in the process Figure 1 each process or multiple processes and / or blocksFigure 1 Apparatus for the functions specified in one or more boxes.

[0065] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction apparatus that implements the functions specified in one Figure 1 process or more processes and / or boxes Figure 1 specified in one or more boxes.

[0066] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one Figure 1 process or more processes and / or boxes Figure 1 specified in one or more boxes.

Claims

1. A multi-scale intelligent simulation method for geothermal development scenarios based on hierarchical coupling, characterized in that, It includes the following steps: S1. According to the geothermal development scenario, obtain geothermal resource data and construct multiple-scale models corresponding to the geothermal resource data; S2. According to the multiple-scale models, construct a multi-scale simulation model, use a deep neural network to perform deep feature learning on the multi-scale simulation model, and obtain the deep feature representation of the multi-scale simulation model; S3. According to the deep feature representation of the multi-scale simulation model, perform feature fusion on the cross-scale feature space of the multi-scale simulation model to obtain a common subspace, and perform feature balancing on the common subspace to obtain a hierarchical coupled multi-scale simulation model. Specifically: S301. According to the deep feature representation of the multi-scale simulation model, perform feature fusion on the cross-scale features of the multi-scale simulation model to obtain a common subspace; Based on the deep feature representation of the multi-scale simulation model, using the learned feature representation of the multi-scale simulation model , where p = 1, 2, …, n; n represents the total number of views, which is the total number of models at different scales, and calculate the covariance tensors of all views , used to discover the correlation information between all views. By adopting simplified calculation, it is transformed into an equivalent problem of maximizing the multi-view canonical correlation, specifically to find a set of optimal rank-one tensors , where = 1, 2, …, r; r represents the feature dimension after dimensionality reduction, which is less than the minimum of the feature dimensions of different views; Make the covariance tensor be representable as a weighted sum of rank-one tensors ; Covariance tensor The calculation expression is as follows: ； Among them, represents the tensor product; Integrate the rank-one tensor into a transformation matrix , and apply the transformation matrix to the feature representation , initially realizing the mapping from the original high-dimensional features to the low-dimensional common subspace; S302. Use the topological manifold structure of the model data to balance the consistent structure and complementary information between the feature representations under different scale models, and obtain the common subspace after balancing the features; S303. Construct a hierarchical coupled multi-scale simulation model through the transformation matrix between the original high-dimensional features and the low-dimensional common subspace; The expression for constructing the hierarchical coupled multi-scale simulation model is as follows: Among them, represents the transformation path, represents the transformation factor, and or 1, represents the transformation matrix from the original high-dimensional feature to the low-dimensional common subspace, represents the transformation matrix from the low-dimensional common subspace to the original high-dimensional feature; S4. Based on the physical process, perform simulation experiment verification on the hierarchical coupled multi-scale simulation model, and use the verified hierarchical coupled multi-scale simulation model to build a simulation platform for the geothermal development scenario to realize multi-scale hierarchical coupled intelligent simulation.

2. The multi-scale intelligent simulation method for geothermal development scenarios based on hierarchical coupling according to claim 1, wherein The S2 includes the following steps: S201. Use the prior knowledge of the same physical process in multiple-scale models and the geothermal resource data to pre-train the physical information neural operator and construct an initial multi-scale simulation model; S202. Perform feature embedding on the initial multi-scale simulation model to obtain the multi-scale simulation model with feature embedding; S203. According to the multi-scale simulation model with feature embedding, design a targeted representation learning model, and according to the targeted representation learning model, construct a deep neural network for learning and representation based on the deep features of the multi-scale simulation model; S204. Use the deep neural network to learn the multi-scale simulation model with feature embedding, learn the feature space of the multi-scale simulation model, and obtain the deep feature representation of the multi-scale simulation model related to the geothermal reservoir.

3. The multi-scale intelligent simulation method for geothermal development scenarios based on hierarchical coupling according to claim 2, characterized in that, The expression for performing feature embedding on the multi-scale simulation model is as follows: Among them, represents the predicted value of the function represented by the neural network at the input coordinates y ; represents the feature embedding of the input function ; represents the feature embedding of the input coordinates y ; q represents the number of basis functions selected in the target space, represents the parameters of the neural network ; Among them, , , represents q d-dimensional Euclidean real space.

Images